Punicalagin Enhances Odontogenic Differentiation and Reparative Dentin Formation via Wnt/β-Catenin Signalling Authors: Yumeng Yang, Xin Li, Qianwen Tang, Chengfei Zhang, Zehao Chen, Wei Liu Published: 2026 ABSTRACT AbstractAimPunicalagin (PCG), a natural polyphenol derived from pomegranate peel, possesses well-established antibacterial and anti-inflammatory properties. However, whether PCG promotes odontogenic differentiation of human dental pulp stem cells (DPSCs) remains unclear. This study investigated the effects of PCG on odontogenic differentiation and reparative dentin formation and further explored the underlying molecular mechanism.MethodsA rat pulpitis model was established to evaluate the effects of a PCG-loaded collagen sponge on pulpal repair. Pulpal inflammation and reparative dentin formation were assessed histologically. DPSCs were isolated and treated with PCG (5 μM). Cell viability, odontogenic differentiation, and Wnt/β-catenin signalling were evaluated using CCK-8, alkaline phosphatase (ALP) and Alizarin Red S staining, quantitative real-time polymerase chain reaction, Western blotting, immunofluorescence, and DKK-1 inhibition assays.ResultsPCG application significantly reduced inflammatory infiltration and stimulated reparative dentin formation in the rat pulpitis model. At the cellular level, PCG enhanced the odontogenic differentiation of DPSCs and upregulated key odontogenic markers (ALP, Runx2, DMP-1, and DSPP) at both the gene and protein levels. Mechanistically, PCG promoted the nuclear translocation of β-catenin, consistent with activation of the canonical Wnt/β-catenin signalling pathway. The pro-odontogenic effects of PCG were largely attenuated by DKK-1–mediated inhibition of canonical Wnt signalling.ConclusionPCG promoted the odontogenic differentiation of DPSCs at least in part through activation of the Wnt/β-catenin signalling pathway. These findings identify PCG as a promising modulator of odontogenic differentiation and provide mechanistic evidence supporting its further investigation for dentin-pulp repair.Clinical relevanceThese findings provide a biological rationale for further investigation of PCG-containing strategies for vital pulp therapy.Key Words: Punicalagin, Dental pulp stem cells, Odontogenic differentiation, Wnt/β-catenin Introduction Dental pulp is a specialised connective tissue enriched with blood vessels and nerves that is indispensable for maintaining tooth vitality and physiological functions.1 Therefore, vital pulp therapy (VPT) is recognised as a preferred biologically based approach for managing pulp injury by preserving the remaining healthy pulp tissue after removal of infected or damaged tissue.2 Compared with nonsurgical root canal treatment, successful VPT preserves the biological functions of the pulp, including immune defence, sensory perception, and reparative capacity, while reducing the susceptibility of treated teeth to fracture.3 The regenerative capacity of the dentin-pulp complex largely depends on dental pulp stem cells (DPSCs), which possess self-renewal and multilineage differentiation capacity.4 Beyond differentiating into odontoblast-like cells, DPSCs can also differentiate into osteoblasts, chondrocytes, adipocytes, and neural cells.5 Following injury induced by bacterial infection, trauma, or thermal irritation, DPSCs are recruited to the injury site, where they proliferate and differentiate into odontoblast-like cells. These newly formed cells subsequently deposit and mineralise dentin matrix, leading to tertiary dentin formation and protection of the underlying pulp tissue.6 Bioactive pulp-capping materials, such as iRoot BP Plus (Innovative BioCeramix), a premixed calcium silicate-based bioceramic, have demonstrated favourable clinical outcomes by promoting reparative dentin bridge formation and improving the success of VPT.7,8 Despite their favourable clinical performance, concerns regarding material cost and handling characteristics have encouraged continued exploration of alternative bioactive strategies. Consequently, naturally derived bioactive compounds and peptide-based therapeutics are being increasingly explored as regenerative candidates owing to their excellent biocompatibility and biological activity. For example, amelogenin-derived peptides have been reported to enhance the proliferation and odontogenic differentiation of human dental pulp cells through activation of the ERK1/2 signalling pathway, highlighting the therapeutic promise of biologically active molecules for dentin-pulp regeneration.9 Among naturally occurring bioactive compounds, punicalagin (PCG), a polyphenolic ellagitannin isolated from pomegranate peel, has emerged as a promising candidate because of its broad spectrum of biological activities and favourable biosafety profile.10,11 Accumulating evidence indicates that PCG exerts antioxidant, antibacterial, anti-inflammatory, and antiviral activities while exhibiting excellent biocompatibility.12, 13, 14 Beyond its anti-inflammatory properties, PCG has also been reported to regulate bone remodelling and attenuate cartilage degeneration through inhibition of MMP-13–mediated type II collagen degradation.15,16 Our previous research demonstrated that PCG alleviates endothelial dysfunction by modulating the NF-κB/VEGFR2/PAK1 signalling pathways.17 More recently, we further showed that PCG alleviates lipopolysaccharide (LPS)–induced inflammatory responses in dental pulp cells through inhibition of the NF-κB/Wnt5a-ROR2 pathway.18 Despite these encouraging findings, it remains largely unknown whether PCG directly regulates the odontogenic differentiation of DPSCs and promotes reparative dentin formation and, if so, through which molecular mechanisms. Controlling inflammation and eliminating bacterial infection are essential goals of VPT. However, successful treatment also requires the regeneration of new dentin to restore the structural integrity and biological function of the tooth.19 Odontogenic differentiation of DPSCs is essential for reparative dentin formation. The canonical Wnt/β-catenin signalling pathway is a critical regulator of tooth development and odontogenic/osteogenic differentiation.20 Kornsuthisopon et al21 showed that tideglusib significantly enhanced the odontogenic/osteogenic differentiation of DPSCs by activating Wnt/β-catenin signalling. Building on previous findings that PCG regulates bone remodelling15 and reduces the inflammatory response in dental pulp,18 we hypothesised that PCG may also promote odontogenic differentiation. Therefore, this study aimed to determine whether PCG promotes odontogenic differentiation of DPSCs and reparative dentin formation and to investigate whether these effects are mediated through activation of canonical Wnt/β-catenin signalling. Methods Characterisation of PCG-loaded collagen sponge PCG-loaded collagen sponges were prepared by incorporating 20 mg PCG (HY-N0063; MedChemExpress) into 10 mL of 1% (w/v) type I collagen solution. The mixture was frozen at −80 °C overnight and lyophilised for 48 hours to generate porous collagen sponges. The dried sponges were then cut into 0.5-mm³ blocks for subsequent animal experiments. To determine the release profile of PCG, an individual PCG-loaded collagen sponge was immersed in 1 mL phosphate-buffered saline (PBS) and incubated at 37 °C under gentle agitation. At each predetermined time point, the release medium was collected and replaced with fresh PBS. The concentration of released PCG was quantified by high-performance liquid chromatography (Agilent Technologies) using an external standard calibration curve, and the cumulative release profile was calculated accordingly. To evaluate sponge degradation, the initial weight (W₀) of each sponge was recorded before immersion in PBS at 37 °C. At the indicated time points, samples were removed, gently blotted to eliminate excess surface liquid, and weighed immediately (Wₜ). The percentage of residual mass was calculated using the following equation: LPS-induced rat pulpitis model All animal procedures were approved by the Animal Ethics Committee of South China Agricultural University (approval No. 2023D079), conducted in accordance with the institutional guidelines, and reported in accordance with the ARRIVE guidelines. Twenty male Sprague-Dawley rats aged 6 to 8 weeks were purchased from the Guangdong Experimental Animal Center (Animal Certificate No. 44007200090387). After 1 week of acclimation, the rats were randomly assigned into 5 groups (n = 4 per group): (1) healthy untreated control, (2) LPS, (3) LPS + blank collagen sponge, (4) LPS + PCG-loaded collagen sponge, and (5) LPS + iRoot BP Plus (Innovative BioCeramix). Blank and PCG-loaded collagen sponges were prepared as described above, sterilised with ethylene oxide, and stored in a desiccator until use. Rats were anesthetised by intraperitoneal injection of 1% pentobarbital. The pulp of the left maxillary first molar was mechanically exposed using a No. 1/4 round steel bur (MANI). Then, 2 μL LPS (10 mg/mL; Sigma-Aldrich) was injected into the pulp chamber to establish pulpitis.22 The exposure site was capped according to the assigned treatment, and all cavities were then restored with glass-ionomer cement (GC). Rats were maintained on a soft diet for 28 days after surgery, and the restorations were inspected daily to confirm their integrity and exclude coronal leakage. Histologic analysis The rats were euthanised 4 weeks after direct pulp capping, and the maxillary jaw segments containing the treated first molars were harvested. Samples were fixed in 4% paraformaldehyde (PFA; Merck) for 24 hours, rinsed under running water overnight, and decalcified in 10% ethylenediaminetetraacetic acid at room temperature for 2 to 3 weeks. Following decalcification, tissues were dehydrated, embedded in paraffin, and sectioned serially at 4 μm (Leica Microsystems). Sections were stained with haematoxylin and eosin (H&E; Boster) and Masson’s trichrome (Beyotime) according to the manufacturer’s instructions. Histologic images were acquired using an upright microscope (ECLIPSE Ci; Nikon). For histomorphometric evaluation, the section exhibiting the largest pulp exposure area and the most representative reparative response was selected from each specimen. Histologic assessment was performed independently by 2 blinded examiners. Pulpal inflammatory response was semiquantitatively evaluated on H&E-stained sections using a modified histologic scoring system adapted from previously published pulp-capping and pulpal healing studies.23, 24, 25 Briefly, inflammation was scored as follows: 1, no inflammatory cell infiltration; 2, mild inflammatory infiltration localised beneath the exposure site; 3, moderate inflammatory infiltration extending into the adjacent pulp tissue; and 4, severe inflammatory infiltration involving a large area of the pulp tissue. In cases of disagreement, a consensus score was assigned following joint evaluation. Quantitative assessment of reparative dentin thickness was performed independently by 2 blinded examiners using ImageJ software (version 1.54d; National Institutes of Health). Following image calibration, measurements were obtained at 3 standardised sites across the newly formed reparative dentin. At each site, the thickness of the reparative dentin was determined as the perpendicular distance between the pulpal surface and the cavity-facing surface of the mineralised tissue. The average of the 3 measurements was calculated and recorded as the reparative dentin thickness for each specimen.26 All values were expressed in micrometres (μm). In cases of disagreement, the final value was determined by consensus. Isolation and characterisation of DPSCs The collection of human dental pulp tissue from extracted teeth was approved by the Ethical Committee of Shenzhen Stomatological Hospital (Pingshan) of Southern Medical University (approval No. 202202A). Healthy premolars and third molars extracted for orthodontic treatment or impaction reasons were collected from 5 donors (3 females and 2 males; aged 15-21 years) at the Maxillofacial Surgery Department of Shenzhen Stomatological Hospital (Pingshan), Southern Medical University. Written informed consent was obtained from all adult donors or, in the case of minors, from both the donor and their legal guardian. All donor teeth were clinically and radiographically healthy, with no evidence of caries, restorations, periodontal disease, or periapical pathology. The extracted teeth were rinsed in PBS (Gibco) supplemented with 2% penicillin-streptomycin (Gibco). The pulp tissue was then digested with type I collagenase for 15 minutes at 37 °C. The digested tissue blocks were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% foetal bovine serum (FBS; Gibco) and 2% penicillin-streptomycin (Gibco) at 37 °C in a humidified atmosphere of 5% CO2. The culture medium was changed every 2 to 3 days. After approximately 7 days, spindle-shaped cells migrated out from the tissue blocks. To assess clonogenic potential, passage 2 cells were subjected to limiting dilution. Single-cell-derived colonies were expanded, fixed in 4% PFA, and stained with 1% crystal violet for 30 minutes to confirm their clonogenicity. Cells from passages 3 to 5 were used for all experiments. To evaluate the multilineage differentiation potential of DPSCs, cells were seeded in 6-well plates at a density of 1 × 105 cells/well and cultured in osteogenic/odontogenic induction medium (DMEM supplemented with 10% FBS, 5 mM β-glycerophosphate, 50 µg/mL ascorbic acid, and 10 nM dexamethasone). The medium was replaced every 3 days. After 7 days, the medium was removed, and the cells were fixed in 4% PFA for 20 minutes before staining with the BCIP/NBT Alkaline Phosphatase (ALP) Color Development Kit (Beyotime). For quantitative analysis, cells were lysed, and ALP activity was measured using an ALP assay kit (Nanjing Jiancheng Biology) with a microplate reader (Tecan). After induction for 14 days, cells were fixed in 4% PFA for 20 minutes and stained with 1% Alizarin Red S (ARS) staining solution (pH 4.2; Solarbio) for 15 minutes. The formed mineral nodules were observed under the microscope (Nikon). For quantitative analysis, the bound dye was dissolved in 100 μL of 10% (w/v) cetylpyridinium chloride (Aladdin) for 20 minutes. Absorbance was measured at 562 nm using a microplate reader (Tecan). For adipogenic differentiation, cells were cultured in adipogenic induction medium (DMEM supplemented with 10% FBS, 0.5 mM 3-isobutyl-1-methylxanthine, 10 μg/mL insulin, 1 μM dexamethasone, and 200 μM indomethacin) for 21 days. Cells were fixed in 4% PFA for 20 minutes and stained with Oil Red O staining solution (Beyotime) for 20 minutes. Surface marker expression was analysed by flow cytometry. DPSCs at passage 2 were incubated with fluorescently conjugated antibodies against human CD29, CD34, CD44, CD45, CD90, CD105, and HLA-DR (Abcam) in the dark for 1 hour at 4 °C. Expression of surface markers was analysed by flow cytometry (Beckman Coulter). Corresponding isotype antibodies were used as controls. Cell viability assay The effect of PCG on DPSC viability was assessed using the Cell Counting Kit-8 (CCK-8; Dojindo). The cells were seeded in 96-well plates at a density of 1 × 103 cells/well. After 24 hours, the cells were exposed to culture medium containing PCG at concentrations ranging from 0 to 50 μM. At 3 and 5 days, CCK-8 reagent was added to each well and incubated for 2 hours. The absorbance was measured at 450 nm using a microplate reader (Tecan). To further evaluate the effect of PCG on DPSC viability, Live/Dead cell staining was performed using a Calcein-AM/Propidium Iodide Cell Viability Assay Kit (Beyotime). At 3, 5, and 7 days, the cells were washed with PBS and incubated with the staining solution at 37 °C for 30 minutes according to the manufacturer’s instructions. Fluorescence images were captured using an inverted fluorescence microscope (IX73; Olympus). Odontogenic differentiation assay DPSCs were seeded in 12-well plates at a density of 5 × 10⁴ cells/well. After 24 hours, cells were cultured in odontogenic induction medium (OM) with or without PCG (5 μM) and/or dickkopf-related protein 1 (DKK-1, 100 ng/mL; MedChemExpress), a Wnt signalling inhibitor, for 7 or 14 days. The total RNA was extracted using the EZ-press RNA Purification Kit (EZBioscience). RNA concentration was determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific). Complementary DNA was synthesised from 1 μg RNA using a Color Reverse Transcription Kit (EZBioscience). Quantitative real-time polymerase chain reaction (qRT-PCR) was performed with SYBR Green qRT-PCR Mix reagent (EZBioscience) on a QuantStudio 6 Flex system (Applied Biosystems). Expression of ALP, Runx2, DSPP, and DMP-1 was normalised to GAPDH and calculated using the 2–ΔΔCt method. Primer sequences are listed in Supplementary Table 1. Protein expression was analysed by Western blotting. Total protein was extracted using protein lysis solution (Beyotime), and the protein concentration was determined using a bicinchoninic acid assay kit (Thermo Scientific). Then, 30 to 50 μg of protein per lane was separated by sodium dodecyl sulphate–polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. Membranes were blocked with 5% skim milk and incubated with primary antibodies against DMP-1 (cat. DF8825, 1:1000; Affinity Biosciences), ALP (cat. DF6225, 1:1000; Affinity Biosciences), Runx2 (cat. AF5186, 1:1000; Affinity Biosciences), DSPP (sc-73632, 1:500; Santa Cruz Biotechnology), and GAPDH (cat. AF7021, 1:5000; Affinity Biosciences) at 4 °C overnight. After incubation with secondary antibodies (cat. SA00001-2, 1:5000; Wuhan Sanying) for 1 hour at room temperature, protein bands were quantified using ImageJ software (version 1.54d; National Institutes of Health). Immunofluorescence staining of DPSCs DPSCs were seeded in 12-well plates at a density of 4 × 10⁵ cells/well. After starvation, cells were cultured in odontogenic induction medium with or without PCG (5 μM) for 3 days. Cells were fixed in 4% PFA, permeabilised with 0.1% Triton X-100 (Beyotime) for 10 minutes, and blocked with 1% bovine serum albumin for 1 hour at room temperature. Cells were incubated overnight at 4 °C with primary antibodies against active β-catenin (cat. 05-665, 1:200; Millipore), followed by incubation with fluorescent secondary antibodies. Nuclei were counterstained with DAPI. Images were captured using a fluorescent microscope (IX63; Olympus). Immunofluorescence staining of pulp tissue Paraffin-embedded sections were deparaffinised, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 6.0). Sections were permeabilised with 0.1% Triton X-100 for 10 minutes and blocked with 1% bovine serum albumin for 1 hour at room temperature. The samples were then incubated overnight at 4 °C with an antiactive β-catenin primary antibody (Millipore), followed by incubation with a corresponding fluorophore-conjugated secondary antibody for 1 hour at room temperature in the dark. Nuclei were counterstained with DAPI. Fluorescence images were acquired using an IX63 microscope. Identical acquisition settings were applied to all experimental groups. Statistical analysis All in vitro experiments were independently repeated using DPSCs derived from at least 3 different donors. For each biological replicate, measurements were performed in triplicate. Data are presented as mean ± standard deviation. Data normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the Brown-Forsythe test before applying parametric statistical analyses. Comparisons between the 2 groups were performed using the Student t test. Multiple-group comparisons were performed using 1-way analysis of variance followed by Bonferroni’s post hoc test. Histologic scores were analysed using the Kruskal-Wallis test followed by Dunn’s multiple-comparison test. Statistical analyses were performed using GraphPad Prism version 10.2.0 (GraphPad Software). A P value <.05 was considered statistically significant (*P < .05, **P < .01, ***P < .001, ****P < .0001). Results Characterisation of the PCG-loaded collagen sponge The in vitro release profile of PCG and the degradation behaviour of the collagen sponge were evaluated to characterise the delivery system. PCG exhibited an initial burst release during the first 48 hours, followed by sustained release, reaching approximately 90% cumulative release by days 8 to 12 (Supplementary Figure 1A). Residual mass measurements showed progressive degradation of both blank and PCG-loaded collagen sponges throughout the observation period (Supplementary Figure 1B). The PCG-loaded collagen sponge degraded slightly faster during the early phase; however, both groups were almost completely degraded by days 10 to 12. PCG attenuates inflammation and promotes reparative dentin formation in vivo To investigate the effects of PCG in vivo, a rat pulpitis model was established, and histologic changes were examined by H&E and Masson’s trichrome staining. The control group exhibited normal pulp morphology with an intact odontoblast layer bordering the dentin (Figure 1A). In contrast, LPS exposure resulted in extensive inflammatory cell infiltration throughout the pulp, accompanied by marked disruption of the normal pulp architecture, focal necrotic changes, and loss of the odontoblast layer. Similar histologic changes were observed in the LPS + blank collagen sponge groups, indicating that the collagen sponge alone did not effectively alleviate pulpal inflammation. Punicalagin (PCG) enhances reparative dentin formation in rat pulp tissue. A, Representative haematoxylin and eosin staining of rat pulp tissue following pulp capping with lipopolysaccharide (LPS), LPS + collagen sponge, LPS + PCG-loaded collagen sponge, and iRoot BP Plus. Insets in the lower region are magnified images of the rectangles. B, Representative Masson’s trichrome staining. The blue-stained areas represent the distribution of collagen fibres (CFs). C, Semiquantitative analysis of pulpal inflammatory response based on histologic inflammation scores. D, Histomorphometric quantification of reparative dentin thickness. Yellow asterisks indicate inflammatory infiltration. RD, reparative dentin. Scale bars: 200 μm (10×) and 100 μm (20×). Data are presented as mean ± SD (n = 4 per group). **P < .01, ****P < .0001. Compared with the LPS and blank collagen sponge groups, the PCG-loaded collagen sponge group exhibited a marked reduction in inflammatory cell infiltration. Mineralised reparative dentin was observed beneath the exposure site, whereas only limited reparative dentin formation was evident in the blank collagen sponge group. The iRoot BP Plus group showed the greatest extent of reparative dentin deposition, with a thicker mineralised barrier and a more continuous odontoblast-like cell layer along the dentin-pulp interface (Figure 1A). Masson’s trichrome staining revealed distinct differences in collagen distribution among the experimental groups (Figure 1B). Sparse collagen deposition was observed in the LPS group, whereas collagen deposition was more prominent in both collagen sponge–treated groups. In the PCG-loaded collagen sponge group, collagen fibres were closely associated with newly formed mineralised tissue beneath the exposure site, whereas only limited mineralised tissue was detected in the blank collagen sponge group. Histologic findings were further quantified by semiquantitative inflammation scoring and histomorphometric measurement of reparative dentin thickness. PCG treatment significantly reduced the inflammatory score compared with both the LPS and blank collagen sponge groups (Figure 1C). Similarly, reparative dentin thickness was significantly greater in the PCG-loaded collagen sponge group than in the blank collagen sponge group, although it remained lower than that observed in the iRoot BP Plus group (Figure 1D). Successful isolation, culture, and characterisation of DPSCs Spindle-shaped, fibroblast-like DPSCs migrated from the tissue explants (Figure 2A). Cells initially exhibited radial outgrowth and formed a vortex-like growth pattern upon reaching 80% confluence (Figure 2B). Limiting dilution produced a homogeneous population of fusiform cells (Figure 2C), which subsequently formed distinct colonies (Figure 2D). Multilineage differentiation was confirmed through functional assays. After 7 days of odontogenic induction, positive ALP staining was observed. ARS staining revealed extensive mineralised nodule formation after 14 days (Figure 2E and F). Following 21 days of adipogenic induction, intracellular lipid droplets were visualised by Oil Red O staining (Figure 2G). Flow cytometry demonstrated a characteristic mesenchymal stem cell immunophenotype, with high expression of CD29 (99.9%), CD105 (99.83%), CD44 (99.96%), and CD90 (99.99%) but minimal expression of CD45 (0.50%), CD34 (0.50%), and HLA-DR (0.48%) (Figure 2H-N). Isolation and characterisation of dental pulp stem cells (DPSCs). A, Primary culture of DPSCs derived from dental pulp tissue. B, Morphology of DPSCs at passage 2. C and D, Representative images of the colony-forming unit-fibroblast (CFU-F) assay demonstrating the clonogenic capacity of DPSCs. E-G, Representative images showing the multilineage differentiation potential of DPSCs, including alkaline phosphatase (ALP) staining (E), Alizarin Red S staining (F) for osteogenic differentiation, and Oil Red O staining (G) for adipogenic differentiation. H-N, Flow cytometric analysis of cell surface markers. DPSCs were positive for mesenchymal stem cell markers (CD29, CD105, CD44, and CD90) and negative for hematopoietic markers (CD45, CD34, and HLA-DR). Scale bars: 500 μm, 200 μm, or 50 μm, as indicated. PCG promotes odontogenic differentiation of DPSCs To determine the optimal concentration and treatment duration of PCG, cell viability was first evaluated at concentrations ranging from 0 to 50 μM after 3 and 5 days of treatment (Figure 3A). At 5 μM, PCG did not significantly affect cell viability after 3 days but significantly reduced cell viability after 5 days (P < .001). Punicalagin (PCG) promotes the odontogenic differentiation and mineralisation of dental pulp stem cells (DPSCs). A, Cell viability of DPSCs treated with various concentrations of PCG for 3 or 5 days, assessed by CCK-8 assay. B, Quantitative real-time polymerase chain reaction analysis of odontogenic-related gene expression, including ALP (7 days) and DSPP (14 days), following pretreatment with selected PCG concentrations for 3 or 5 days. C and D, Representative images of alkaline phosphatase (ALP) staining (7 days) and Alizarin Red S (ARS) staining (14 days). E, Quantitative analysis of ALP activity and ARS mineralisation. Scale bars: 500 μm. *P < .05, ** P < .01 vs control. ns, not significant. For odontogenic differentiation, cells were treated with PCG-containing induction medium for the initial 3 or 5 days, followed by PCG-free induction medium until 7 days (ALP) and 14 days (DSPP) (Figure 3B). Treatment with 5 μM PCG for 3 days significantly increased ALP and DSPP expression without affecting cell viability. Although 5-day treatment resulted in slightly higher ALP and DSPP expression, it was also associated with significantly reduced cell viability. To further evaluate the persistence of PCG-associated cytotoxicity, DPSC viability was assessed under odontogenic induction conditions. Continuous exposure to 5 μM PCG (maintenance group) significantly reduced cell viability at 5 and 7 days compared with the OM group (Supplementary Figure 2A). In contrast, cells in the withdrawal group (3-day exposure followed by PCG-free culture) showed viability comparable to the OM group at both time points, with no significant differences (Supplementary Figure 2A). Live/Dead staining confirmed these findings, showing mainly viable cells in the withdrawal and OM groups, whereas increased cell death was observed in the maintenance group (Supplementary Figure 2B). These results support that 3-day exposure to 5 μM PCG does not induce persistent cytotoxicity and is suitable for subsequent experiments. PCG significantly enhanced both early and late odontogenic differentiation markers. ALP staining showed more intense signals in the PCG group than in controls (Figure 3C), consistent with increased ALP activity (Figure 3E). At day 14, ARS staining and semiquantitative analysis revealed increased mineralised nodule formation in the PCG group (Figure 3D and E). These results demonstrate that PCG promotes odontogenic differentiation and matrix mineralisation. PCG upregulates odontogenic marker expression qRT-PCR analysis showed that the mRNA expression levels of ALP, Runx2, DSPP, and DMP-1 were significantly higher in the PCG-treated group than in the control group under odontogenic induction conditions (Figure 4A-D). Consistently, Western blot analysis demonstrated that the protein levels of ALP, Runx2, DSPP, and DMP-1 were also significantly upregulated following PCG treatment (Figure 4E and F) (P < .05). Punicalagin (PCG) enhances the odontogenic differentiation of dental pulp stem cells (DPSCs) under odontogenic induction. A-D, Quantitative real-time polymerase chain reaction analysis of odontogenic-related gene expression, including ALP (A), Runx2 (B), DSPP (C), and DMP-1 (D), in DPSCs cultured in odontogenic induction medium (OM) with or without PCG. E, Western blot analysis showing the protein levels of DMP-1, ALP, Runx2, and DSPP, with GAPDH as the loading control. F, Quantitation of the relative protein expression levels normalised to GAPDH. Data are presented as mean ± SD (n = 3). *P < .05, **P < .01 vs the OM group. PCG promotes odontogenic differentiation via activation of the Wnt/β-catenin pathway The Wnt/β-catenin signalling pathway is a key regulator of odontogenic differentiation in DPSCs.27,28 To delineate the underlying mechanisms of PCG-mediated effects, the involvement of this pathway in both cellular and in vivo models was further investigated. Immunofluorescence staining of active β-catenin, the downstream effector of canonical Wnt signalling,29 revealed predominantly cytoplasmic localisation under odontogenic induction conditions. In contrast, PCG treatment increased β-catenin fluorescence intensity and promoted its nuclear translocation in DPSCs, indicating activation of Wnt/β-catenin signalling (Figure 5A). Punicalagin (PCG) promotes odontogenic differentiation via the Wnt/β-catenin signalling pathway. A, Immunofluorescence staining of active β-catenin (red) in dental pulp stem cells (DPSCs). White arrows indicate the nuclear translocation of active β-catenin; nuclei were counterstained with DAPI (blue). Scale bars: 20 μm. B and C, Representative images of alkaline phosphatase (B) and Alizarin Red S (C) staining in DPSCs under different treatments (odontogenic induction medium [OM], OM + PCG, OM + DKK-1, OM + DKK-1+ PCG). D, Quantitative real-time polymerase chain reaction analysis of ALP, Runx2, DSPP, and DMP-1 messenger RNA levels. E, Western blot analysis of DMP-1, ALP, Runx2, and DSPP protein levels. F and G, Quantitative analysis of relative protein levels normalised to GAPDH. H, Immunofluorescence staining of active β-catenin in rat dental pulp tissue. The white line indicates the boundary between original dentin and pulp tissue, while the red line indicates the interface between newly formed reparative dentin and underlying pulp tissue. Scale bars: 50 μm. n = 4. *P < .05, **P < .01, ***P < .001, ****P < .0001 vs OM group; #P < .05 vs OM + PCG group. ns, not significant. To confirm functional dependence, cells were cotreated with the Wnt inhibitor DKK-1. DKK-1 significantly attenuated the PCG-induced ALP activity (Figure 5B) and mineralised nodule formation (Figure 5C). Consistently, PCG-driven upregulation of odontogenic markers (ALP, Runx2, DSPP, and DMP-1) at both messenger RNA and protein levels was largely reversed by Wnt inhibition (Figure 5D-G). To validate in vivo pathway activation, immunofluorescence staining was performed on rat pulp tissues. In the control and iRoot BP Plus groups, active β-catenin was mainly localised in the cytoplasm of odontoblast-like cells at the pulp-dentin interface, with limited nuclear staining. In contrast, the PCG-loaded collagen sponge group showed marked nuclear accumulation of active β-catenin in odontoblast-like cells beneath the reparative dentin bridge, indicating enhanced Wnt/β-catenin activation in vivo (Figure 5H). These findings preliminarily support the involvement of canonical Wnt/β-catenin signalling in the pro-odontogenic effects of PCG. Discussion VPT has gradually evolved from a treatment focused on preserving pulp vitality to a biologically based strategy that promotes regeneration of the dentin-pulp complex.30 Accordingly, increasing attention has been given to bioactive molecules that can both modulate the inflammatory microenvironment and stimulate the regenerative potential of endogenous DPSCs.31 In our study, local application of PCG significantly attenuated pulpal inflammation and promoted reparative dentin formation in a rat pulpitis model. These in vivo findings were supported by our in vitro results, which demonstrated that PCG enhanced the odontogenic differentiation of human DPSCs and promoted mineralised matrix formation. These findings expand the biological profile of PCG beyond its previously reported anti-inflammatory and antioxidant activities by identifying its direct pro-odontogenic effects on DPSCs. More importantly, they further support the therapeutic potential of PCG for dentin-pulp repair. Because DPSCs are the principal mesenchymal stem cells responsible for tertiary dentin formation following pulp injury, strategies that simultaneously establish a favourable inflammatory microenvironment and promote DPSC differentiation are considered fundamental for successful pulp regeneration.19 Taken together, these findings identify PCG as a promising naturally derived bioactive molecule for dentin-pulp regeneration and provide a biological basis for its further investigation in VPT. This concept is further supported by recent evidence showing that modulation of key signalling pathways can improve dental tissue regeneration. For example, activation of the AKT/endothelial nitric oxide synthase axis restored impaired dentin regeneration in diabetic teeth, highlighting the therapeutic potential of pathway-based strategies for regenerative endodontics.32 To further elucidate the molecular mechanism underlying the pro-odontogenic effects of PCG, we investigated the involvement of the canonical Wnt/β-catenin signalling pathway, which is recognised as a central regulator of odontoblast differentiation and dentin regeneration.33 In the present study, PCG promoted the nuclear translocation of active β-catenin in DPSCs, indicating activation of canonical Wnt signalling. Moreover, inhibition of this pathway with DKK-1 markedly attenuated PCG-induced ALP activity, mineralised matrix formation, and the expression of odontogenic markers. These findings suggest that activation of the Wnt/β-catenin pathway is one of the mechanisms contributing to the pro-odontogenic effects of PCG. Similar findings have been reported for several bioactive molecules that enhance odontogenic differentiation through activation of canonical Wnt/β-catenin signalling. For example, de Souza Araújo et al34 demonstrated that local delivery of Wnt proteins promoted odontogenic differentiation and dentin-pulp regeneration, further supporting an important role for Wnt signalling during dental tissue regeneration. Although the bioactive agent differs from PCG, these findings provide further evidence that activation of canonical Wnt/β-catenin signalling is a common mechanism underlying bioactive molecule-induced odontogenic differentiation. Collectively, these studies, together with our findings, support an important role for canonical Wnt/β-catenin signalling in regulating odontogenic differentiation. Interestingly, although iRoot BP Plus induced the greatest reparative dentin formation in our study, active β-catenin remained predominantly localised in the cytoplasm. Previous studies have shown that calcium silicate–based bioceramic materials promote dentin-pulp repair through multiple biological mechanisms, including bioactive ion release and stimulation of mineralisation-related cellular responses, suggesting that PCG and iRoot BP Plus may facilitate reparative dentin formation through partially distinct molecular mechanisms.35 However, it should be acknowledged that, although DKK-1 significantly attenuated the pro-odontogenic effects of PCG in our study, it did not completely abolish them. This finding suggests that activation of the canonical Wnt/β-catenin signalling pathway alone cannot fully explain the biological effects of PCG. Rather than acting through a single signalling cascade, PCG may exert its proregenerative effects through coordinated regulation of multiple molecular pathways. This interpretation is further supported by our previous study, which demonstrated that PCG alleviates inflammatory responses in dental pulp cells through modulation of the NF-κB/Wnt5a-ROR2 signalling pathway.18 In addition, accumulating evidence indicates that the anti-inflammatory and antioxidant activities of PCG are closely associated with the regulation of oxidative stress and inflammatory mediators.36,37 Therefore, PCG is likely to exert its biological effects through coordinated regulation of multiple interconnected signalling pathways rather than a single canonical Wnt-dependent mechanism.38 Importantly, although our in vitro findings strongly support the involvement of Wnt/β-catenin signalling, direct functional validation of this pathway in vivo was not performed in the present study. Given the complexity of the pulp microenvironment, pharmacologic or genetic modulation of Wnt/β-catenin signalling in animal models would provide more definitive evidence for its contribution to PCG-mediated pulp repair. Such studies will also help clarify the interplay between canonical Wnt signalling and other inflammation-related pathways during dentin-pulp regeneration. Based on the previously reported anti-inflammatory properties of PCG and its regenerative potential in other tissue repair models,15,39 we hypothesised that PCG might also promote dentin-pulp regeneration. Accordingly, the primary objective of the present in vivo study was not to develop a clinically applicable pulp-capping material but rather to investigate the biological effects of locally delivered PCG on pulp repair. To minimise potential confounding effects of the delivery vehicle, a collagen sponge was selected as a simple carrier because of its excellent biocompatibility, biodegradability, and widespread use as an experimental scaffold for the local delivery of bioactive molecules in dental tissue engineering and regenerative endodontics.40,41 Consistent with this rationale, the blank collagen sponge alone produced only limited improvement in pulpal healing, whereas incorporation of PCG markedly reduced inflammatory infiltration and promoted reparative dentin formation. These findings suggest that the observed therapeutic effects were primarily attributable to PCG, while the collagen scaffold mainly functioned as a biocompatible matrix for local drug delivery. Nevertheless, the collagen sponge used in this study should not be considered a clinically optimised pulp-capping material. Although it provided an appropriate platform for evaluating the biological activity of PCG, further optimisation of the delivery system will be required before clinical translation. In particular, incorporation of PCG into clinically applicable pulp-capping materials should be accompanied by comprehensive evaluation of release kinetics, sealing ability, degradation behaviour, mechanical properties, antibacterial performance, and long-term biocompatibility. Therefore, the present findings should be interpreted as biological proof-of-concept evidence supporting the regenerative potential of PCG, rather than as direct validation of a clinically applicable pulp-capping strategy. Despite the encouraging findings, several limitations should be considered when interpreting the present results. First, although the combined use of a rat pulpitis model and human DPSCs provided complementary evidence supporting the biological activity of PCG, species-specific differences in pulp biology, immune responses, and tissue repair should be considered when extrapolating these findings to human clinical conditions.42 In addition, the relatively short observation period enabled assessment of the early reparative response but did not allow evaluation of the long-term maturation, structural stability, or functional performance of the newly formed reparative dentin. Likewise, although statistically significant differences were observed, the exploratory nature of this proof-of-concept study and the relatively small animal sample size warrant further validation in larger animal cohorts. Future studies incorporating longer follow-up periods and clinically relevant preclinical models will provide more robust evidence regarding the durability, long-term biological safety, and translational potential of PCG for vital pulp therapy. Taken together, the present findings establish a biological foundation for future investigations aimed at optimising PCG-based therapeutic strategies and further evaluating their long-term efficacy and regenerative performance in clinically relevant settings. Conclusion Our results demonstrate that PCG promotes the odontogenic differentiation of DPSCs and enhances reparative dentin formation. These effects are at least partially associated with the activation of the Wnt/β-catenin signalling pathway. Collectively, these findings identify PCG as a promising bioactive molecule worthy of further investigation for VPT. Author contributions Yumeng Yang conceived and designed the study, conducted the experiments, analysed the data, and drafted the manuscript. Xin Li, Qianwen Tang, Chengfei Zhang, and Zehao Chen contributed to data acquisition, analysis, and interpretation. Wei Liu conceived and supervised the project. All authors critically reviewed, revised the manuscript for important intellectual content, and approved the final version to be submitted. Funding This research was supported by the Special Innovation Projects for General Universities in Guangdong Province (No. 2021KTSCX016) and General Projects of the President’s Scientific Research Fund of Shenzhen Stomatological Hospital of Southern Medical University (Pingshan) (Nos. 2021B002 and 2022B003). Technology and Innovation Commission of Shenzhen Municipality (Grant No. JCYJ2023080715430410). Data availability The datasets generated are available from the corresponding author upon reasonable request. Conflict of interest All authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article. Footnotes Supplementary material associated with this article can be found in the online version at doi:10.1016/j.identj.2026.109801. Contributor Information Yumeng Yang, Email: yumengyang88@sohu.com. Xin Li, Email: xli0712@hku.hk. Qianwen Tang, Email: qwtang666666@163.com. Chengfei Zhang, Email: zhangcf@hku.hk. Zehao Chen, Email: chenzehaoscience@163.com. Wei Liu, Email: lenmon1@126.com. Supplementary materials Supplementary Fig. 1 Characterisation of the punicalagin (PCG)–loaded collagen sponge. A, Cumulative release profile of PCG from the PCG-loaded collagen sponge in phosphate-buffered saline at 37 °C. B, Degradation profiles of blank collagen sponges and PCG-loaded collagen sponges, expressed as the percentage of remaining mass over time. Data are presented as mean ± SD (n = 3).*P < .05, **P < .01. ns, not significant. Supplementary Fig. 2 Evaluation of dental pulp stem cell (DPSC) viability following transient or continuous exposure to punicalagin (PCG). A, CCK-8 assay of DPSC viability at days 3, 5, and 7. OM, odontogenic medium; Withdrawal, 5 μM PCG applied for the initial 3 days only; Maintenance, continuous exposure to 5 μM PCG. B, Representative Live/Dead staining images. Live cells are shown in green and dead cells in red. Scale bars: 200 μm. Data are presented as mean ± SD (n = 3). *P < .05, ***P < .001. ns, not significant. References Associated Data Supplementary Materials Supplementary Fig. 1 Characterisation of the punicalagin (PCG)–loaded collagen sponge. A, Cumulative release profile of PCG from the PCG-loaded collagen sponge in phosphate-buffered saline at 37 °C. B, Degradation profiles of blank collagen sponges and PCG-loaded collagen sponges, expressed as the percentage of remaining mass over time. Data are presented as mean ± SD (n = 3).*P < .05, **P < .01. ns, not significant. Supplementary Fig. 2 Evaluation of dental pulp stem cell (DPSC) viability following transient or continuous exposure to punicalagin (PCG). A, CCK-8 assay of DPSC viability at days 3, 5, and 7. OM, odontogenic medium; Withdrawal, 5 μM PCG applied for the initial 3 days only; Maintenance, continuous exposure to 5 μM PCG. B, Representative Live/Dead staining images. Live cells are shown in green and dead cells in red. Scale bars: 200 μm. Data are presented as mean ± SD (n = 3). *P < .05, ***P < .001. ns, not significant. Data Availability Statement The datasets generated are available from the corresponding author upon reasonable request.